In general, in the case that ambient noise is severe in two or more multi-party conversations, the strength of speech of a speaker that is transferred to the listener's ears is reduced in inverse proportion to the square of the distance to reach the listener's ears. Thus, in the case ambient noise of the listener is still severe, it is difficult for the listener to hear the voice of the speaker. But, since the voice of the speaker reaches from the mouth to the ears, in the range of 20 cm or less, the speaker may hear his or her own voice clearly even if the ambient noise is great.
Thus, if the same principle as described above is applied to the listener and conversation partners are able to hear voice translation in real-time in two or more multi-party conversations, regardless of the languages spoken by the two or more conversation partners, the conversation partners may clearly recognize dialogues between them even under the severe noisy circumstances around each other.
The present inventor has noted that if a technology of an ear implantable, that is, an ITE (In-The-Ear) type radio transceiver, a technology of a repeater in which a central processor (CPU) is housed as in smart phones to thus process digital audio signals, a speech recognition technology, and a voice translation technology are integrated with each other, speakers and listeners such as persons with normal hearing, persons with hearing loss, and people who speak different languages, may converse with each other even in noisy surroundings without hindrance.